Klaus Seidel

Toward Snowmelt Runoff Forecast Based on Multisensor Remote-Sensing Informnation

Snow-cover mapping is a prerequisite for deriving the main input variable for a deterministic snowmelt runoff model (SRM). Operational forecasting of snowmelt runoff has to rely on guaranteed snow cover information for a specific area and time. Remote sensing can provide the necessary data, especially when different sensor systems are available for combined interpretations. It has been shown already how remote-sensing information from Landsat-MSS and NOAA/AVHRR supplement each other. It is important, though, that when using coarse spatial sensor resolution, special attention to the area, its topography, and the amount of snow coverage must be taken.

Augmenting LANDSAT MSS Data with Topographic Information for Enhanced Registration And Classification

This paper deals with problems arising in the classification of LANDSAT MSS data from rugged terrain. A digital terrain model (DTM) was found to be useful in several ways. For registration by cross-correlation, mountain ridges were extracted from both a synthetic image based on the DTM and a LANDSAT image. Information from the DTM, from thematic maps, and meteorological data were all used as ancillary data to aid in rapid snow cover determination without direct ground control in a large catchment area. In addition it is shown that the use of the DTM not only allows the assessment of relative and absolute snow distribution within given elevation zones, but also permits the extrapolation of snow cover into areas partly covered with clouds.

Large-Area Deterministic Simulation of Natural Runoff from Snowmelt Based on Landsat MSS Data

This work sentnow a method of periodic evaluation of snow-covered areas by digital processing of Landsat data. Since the s cover was mapped for the first time in a large and morphologicomplex alpine basin, it was necessary to develop procedures to determine the snow coverage for partly clouded regions or for incomplete satellite scenes. The changing areal extent of the seasonal snow cover is an important variable for deterministic snowmelt runoff models. By using the SRM model, the natural runoff in the Rhein-Felsberg basin (3249 kM, 571-3614 m a.s.I) was simulated although the measured river flows are significantly influenced by artificial reservoir operation. Such simmulation would not be possible by calibration models that optimize the model parameters by the measured discharge.

An innovative concept for image information mining

Information mining opens new perspectives and a huge potential for information extraction from large volumes of heterogeneous images and the correlation of this information with the goals of applications. n nWe present a new concept and system for image information mining, based on modelling the causalities which link the image-signal contents to the objects and structures within interest for the users. The basic idea is to split the information representation into four steps: n n1. image feature extraction using a library of algorithms such to obtain a quasi-complete signal description n n2. unsupervised grouping in a large number of clusters to be suitable for a large set of tasks n n3. data reduction by parametric modelling the clusters n n4. supervised learning of user semantics, that is the level where, instead of being programmed, the systems is trained by a set of examples; thus the links from image contents to the users are created. n nThe record of the sequence of links is a knowledge acquisition process, the system memorizes the user hypotheses. Step 4. is a man-machine dialogue, the information exchange is done using advanced visualization tools. The system learns what the users need. n nThe system is presently prototyped for inclusion in a new generation of intelligent satellite ground segment systems, value adding tools in the area of geoinformation, and several applications in medicine and biometrics are also forseen.

Experiences from real time runoff forecasts by snow cover remote sensing

Seasonal and short-term runoff forecasts for two hydroelectric stations on the upper Rhine basin are carried out in real time based on snow cover monitoring by Landsat and SPOT satellites. Evaluation of snow reserves on 1 April 1993 from satellite data reveals uncertainties in estimates using point measurements on the ground as index. Runoff is computed by the SRM model with snow covered areas as well as temperature and precipitation forecasts as input variables. These runoff forecasts can be exploited, among other purposes, for optimizing the hydropower production and for timely decisions on the electricity market.<<ETX>>

Alpine snow cover and glaciers in relation to altitude from advanced satellite monitoring

Advanced methods of Landsat-TM data processing were used to map periodically the seasonal snow cover and glaciers. The test basins are located in western, eastern and southern Swiss Alps. In the basins of the rivers Rhone at Sion (3371 km/sup 2/, 488-4634 in a.s.l.), Rhine at Felsberg (3249 km/sup 2/ 562-3614 in a.s.l.) and Ticino at Bellinzona (1515 km/sup 2/, 220-3402 in a.s.l.) relations were derived between the snow coverage and altitude at different stages of the snowmelt season. From these curves, the snow coverage at a desired altitude and the elevation of the statistical snow line at different dates can be read off. The decrease of the areal extent of snow on glaciers is slower in comparison with glacier-free areas. The snow line in the basin Rhone-Sion is lower than in the other basins throughout the season. Precision snow cover mapping and an adequate evaluation of data are needed for runoff modelling and winter tourism, in particular with regard to global warming.

Information theoretical assessment of methods for segmentation of high resolution remote sensing images

Scene understanding of remotely sensed images requires a certain amount of preprocessing in order to remove, or alleviate the effects of, all those factors that disturb the imaging process. These factors depend essentially on the peculiar way in which each kind of sensor acquires the image (sensor-related factors) and on the terrain topography, the illumination and the view angle (radiometric factors). In this paper, a Bayesian model-based maximum a posteriori estimation approach to correct these disturbing factors is suggested.